RU2574040C2 - Method and device for determining probability of exiting area identified on digital map as open terrain - Google Patents

Abstract

FIELD: radio engineering, communication.

SUBSTANCE: method of determining the probability of exiting an area identified on a digital map as open terrain, starting from a position within the area, includes dividing the area into a plurality of cells, with an initial transition probability being assigned to one transition from one cell to an adjacent cell, between the position and the edge of the area; adjusting the initial transition probability, performed in the presence of a type of information from the digital map relating to a cell or adjacent cell to obtain an adjusted transition probability and determining the exit probability, performed using the adjusted transition probability. The invention also relates to an apparatus for determining the probability of exiting an area according to said method.

EFFECT: faster determination or plotting of a route.

9 cl, 5 dwg

Description

FIELD OF THE INVENTION

The present invention relates to a method for determining at least one probability of leaving an area identified in an digital map as an open area, to a device for determining at least one probability of leaving an area identified in a digital map as an open area, as well as to a corresponding computer software product.

State of the art

Known navigation systems based on road networks form a route proposal, or a proposed route, based on the information base laid down in them. As a rule, such an information base is a geographic digital map, which includes roads. The proposed route can be transmitted through the communication system to the driver of the vehicle as a recommendation for movement. At the same time, by determining the location of the vehicle, a specific location of the vehicle and the route proposed for it are negotiated, and, taking into account the situation in which the vehicle is located, as well as with pre-emption regarding the proposed route, traffic recommendations are developed. Since the result of the location determination is not an absolute, but only a probable location, it is necessary that the roads contained in the digital map be widened on both sides by a certain tolerance area. If the vehicle’s specific location is within the tolerance range, then the vehicle is considered to be on the road. This method, called the map snap method (map matching), indicates that the current location of the vehicle is off the road, only when it goes beyond the specified tolerance area. In this case, the known method cannot correlate with the current location of the vehicle any road or adjoining area of tolerance and characterizes the vehicle as being on an off-road or open terrain. After the vehicle, having left the non-road area, is physically on the road along which it will continue to travel, the known method for linking to the map will take some time until the vehicle is identified as being on the road, and the adapted driver can again be issued with an adapted to a changed situation, a recommendation for movement.

DE 10146115 B4 describes a method for navigating a vehicle to a destination by navigating with a direction of travel if the vehicle is identified as being outside a known road network and in the absence of an alternative known road that could bring the vehicle to its destination.

Disclosure of invention

Based on the foregoing, in the framework of the present invention, a method for determining at least one probability of leaving an area identified in the digital map as an open area, a device for determining at least one probability of leaving an area identified in a digital map as an open area, and the corresponding computer program product described in the independent claims. Particular embodiments of the invention are given in the respective dependent claims, as well as discussed in the description below.

The present invention is based on the realization that a time delay in starting a route after reaching a known road leading from an open area can be avoided by determining in advance the probability of a vehicle leaving an open area, namely from the current location of the vehicle in an open area. This allows you to prepare route guidance in advance, having determined the probable exit point from the open area, located on the road adjacent to this open area. To achieve a higher probability of predicting an open area and its limiting objects, you can associate attributes from a digital map that increase or decrease the likelihood of driving through a certain area of the terrain or reaching a certain point on the terrain. Based on these attributes, it is possible, based on the current location of the vehicle in an open area, to determine in advance the probability of reaching one or more points on an object bounding an open area (restricting an open area). In addition, with this determination, other factors can be taken into account, such as the current orientation of the vehicle or the current speed of the vehicle.

An advantage of the invention is that by applying the method corresponding to the approach discussed here, it is possible to quickly determine or get directions before the vehicle leaves the area identified as an open area. Also, thanks to the invention, the driver of the vehicle will be able to more quickly follow the recommendations for movement and turns on the route and faster to reach a given destination. By eliminating errors in determining the route at the end of leaving the open area, you can minimize the distance traveled and avoid excessive energy consumption.

One object of the present invention is a method for determining at least one probability of leaving the area identified in the digital map as an open area based on the location within the area, comprising the following steps:

- dividing the region into many cells with assigning at least one transition between the location and the edge of the region from one cell to the neighboring cell of the initial probability of transition;

- correction of the initial transition probability, performed in the presence of at least one type of information from a digital map relating to a cell or an adjacent cell, with obtaining an adjusted transition probability; and

- determination of the probability of exit, performed, at least, using the adjusted probability of transition.

The object of the present invention is also a device for determining at least one probability of leaving the area identified in the digital map as open area, based on the location within the area, containing:

- a node for dividing a region into many cells with assigning at least one transition between one location and the edge of the region from one cell to the neighboring cell of the initial probability of transition;

- a node for correcting the initial probability of the specified transition, performed when there is at least one type of information from the digital map related to the specified cell or a cell adjacent to it, with obtaining the adjusted probability of transition; and

- a node for determining the probability of exit, which is performed at least using the adjusted transition probability.

A digital map may be a representation of the collected data with reference in the spatial coordinate system. The data collected may represent space phenomena. For example, the phenomena of space may be communication lines, terrain, water bodies or vegetation. Space phenomena can be stored in a digital map in a generalized and digital form. Land plots without communication lines can be indicated in a digital map as an open area or an off-road area. An open area may have its own characteristics. An area identified as an open area may be limited by the spatial phenomena covered by the map, such as communications, settlements or bodies of water. The area may be subdivided, or divided into sections. Partitioning can be performed using arbitrary or systematically set boundaries. Borders can divide a region into cells. Cells can be arranged orderly or disordered. Cells can have any shape on the plane, such as, for example, triangular, quadrangular, pentagonal or hexagonal, etc. Two neighboring cells can have a transition at their common border. Two cells with a single point of contact can have a transition, for example, at a nodal point where several boundaries converge. The transition from the first cell to the second cell can be assigned the initial probability of transition. In the first cell, all the initial probabilities of the transition to neighboring cells can be combined to obtain a common, or total, value of the transition probabilities, for example, a value equal to one. The initial transition probabilities may be equal. The initial transition probability can be adjusted by following the steps of the method of the invention. At the correction step, the value of the initial transition probability can be changed, i.e. the value of the initial transition probability can be increased or decreased. As a result of the correction, the adjusted transition probability is obtained. Increasing or decreasing the value may represent a greater or lesser likelihood of leaving or reaching the corresponding cell. This probability can be affected by the phenomena of cell space or a cell adjacent to it. These space phenomena can be displayed with information contained in a digital map. For example, it is unlikely that a vehicle will reach a terrain that is difficult for such a vehicle. Then the probability of reaching an impassable stretch of terrain may tend to zero. Accordingly, it is more likely that the vehicle will reach a well-traveled area or a vehicle will travel over such terrain, while at the same time, it will be less likely to leave a well-traveled area. By means of adjusted transition probabilities, and if correction is not possible, by means of initial transition probabilities, it is possible to determine the exit probability for a cell located at the edge of an open area / off-road area. For this, the probabilities of transitions between a location located within the region and a cell located at the edge of the region can be combined according to specific processing instructions. Such a processing instruction may be, for example, a Markov chain, each element of which may represent one corresponding cell, and Markov probabilities between elements of the chain may represent transition probabilities.

The present invention also relates to a device or a navigation device made to implement or implement the steps of the method of the invention. In particular, such a device or navigation device may have (functional) nodes made to implement the corresponding steps of the method. When implementing the invention in this embodiment, namely in the form of a control unit, the underlying problem is also solved quickly and efficiently.

A device or a navigation device in this application can be understood as an electric device that processes signals from sensors and, depending on them, provides navigation signals. The navigation device may have an interface that can be implemented in hardware and / or software. In the case of hardware implementation, the interfaces can be, for example, part of the so-called system specialized integrated circuit (ASIC), which includes a variety of functions of the navigation device. At the same time, it is also possible that the interfaces are their own integrated circuits or at least partially consist of discrete components. In the case of a software implementation, the interfaces can be software modules contained, for example, in a microcontroller along with other software modules.

It is also preferable to carry out the invention in the form of a computer program product with program code recorded on a computer-readable storage medium, such as a semiconductor memory device, a hard disk drive or an optical drive, and used to carry out the method corresponding to one of the above embodiments of the invention with running a program in a device or navigation device.

In one embodiment of the present invention, in a split step, several transitions may form a network, the network covering the area and / or cells limited to a predetermined maximum area of the area. The network can cover the area completely and continuously. If the size, or area, of a region exceeds a predetermined value, the region may be limited by one or more auxiliary boundaries to a predetermined value. For example, a given value may be one square kilometer. This allows you to limit the processing time for performing the steps of the method and perform the steps of the method quickly.

In another embodiment of the present invention, in the correction step, the initial transition probability can be adjusted using the class of the message path if the cell includes at least this message path and the digital map contains information about the message path and its class, and / or in which the initial transition probability can be adjusted using at least one terrain characteristic in a cell or in an adjacent cell if the digital map contains information about the terrain characteristic. The message path class can represent certain information regarding the probability of reaching and / or leaving the message path. For example, the probability of moving from an adjacent cell to a cell with a low-class road, such as a country or local road, may be higher than the probability of moving into a cell with a higher-class road, such as a road for high-speed traffic. The probability of going into a cell with a high-class road, such as a expressway, may tend to zero. You can get to the expressway, as a rule, exclusively through the corresponding entrances, i.e. from adjoining roads of smaller classes. Thus, it is advisable that the determined probability of reaching a cell containing a very high-class road, for example, a expressway, be very small or equal to zero. The terrain can also affect the cell’s traffic. For example, the probability of a transition from a cell with a small slope of the terrain to a cell in which there is no terrain slope can be higher than the probability of a transition from a cell with a small slope of the terrain to a cell with a strong slope of the terrain. The probability of transition to cells that are impassable, i.e. unsuitable for vehicle movement, due to the presence of characteristics or features such as cliffs, steep slopes, water surfaces or dense forest, as well as development, may tend to zero. If impassable cells limit the area, then it is advisable that the probability of leaving the area through such cells is as low as possible. This eliminates the possibility of incorrect calculation of offers along the route and takes into account for preliminary calculation only promising exit points from an open area.

In yet another embodiment of the present invention, in the correction step, the initial transition probability can be adjusted using the instantaneous direction of movement and / or the instantaneous speed of the vehicle in which the method of the invention is implemented. Thus, in determining the probability of an exit, the instantaneous (current) state of movement of the vehicle can be taken into account. For example, it is unlikely that the vehicle will make extreme changes in direction and / or speed. The vehicle’s instantaneous speed can affect its ability to corner and bend roads. For example, when moving at the speed of a pedestrian, the vehicle can go through steeper turns than when driving at maximum speed. Thus, the likelihood of transitions located directly at the heading of the vehicle, or transitions achievable at moderate steering angles, may be higher than the probability of transitions located, for example, next to the vehicle or opposite to the instantaneous direction of movement, i.e. back at the heading.

Further, in yet another embodiment of the invention, in the correction step, the initial transition probability can be adjusted using the transition distance to the vehicle’s entry point to an open area. For example, a parking spot for vehicles or entry into a privately owned territory may not be included in the map and thus may be laid out in the map as an open area. Then you can expect that the vehicle will leave the open area in the same place where it entered the open area. If the vehicle is at a very great distance from the entry point, then it may be more likely that the vehicle will leave the off-road area / open area elsewhere.

In yet another embodiment of the invention, in the step of dividing at least one other transition between the cell and the edge of the region between the cell to another neighboring cell, a different initial transition probability or an initial probability of another transition can be assigned. In the step of correcting the initial probability, the other initial probability of the transition can be adjusted if at least one other type of information from the digital map relating to the cell or another neighboring cell is available to obtain an adjusted different transition probability. In the determination step, a different exit probability can be determined, the other exit probability being determined at least using an adjusted different transition probability. One transition may be between the first and second cells. Another transition may be between the first and third cells. These transitions are between cells directly connected to each other, i.e. between cells reached directly from each other. The third cell may be located in close proximity to the first cell. The second and third cells are not necessarily directly adjacent to each other, but can be located in this way. Another transition can be assigned a different (own) initial probability of transition. This other initial transition probability, or the initial probability of another transition, can be called the second initial transition probability. The correction of the other initial transition probability can be carried out in accordance with the information obtained from the digital map in relation to the first and / or third cell. The adjusted other transition probability can be called the second adjusted transition probability, and it will characterize the transition between the first and third cells. The probability of leaving the area may be the first probability of leaving. Another probability of leaving the area may be the second probability of exit. The first and second exit probabilities can relate to different cells, but both probabilities can also relate to the same cell, however when using different transition probabilities. The first transition probability can be used to determine the first exit probability, the second transition probability can be used to determine the second exit probability. A similar embodiment of the present invention provides the advantage that it allows you to determine the most likely exit point from the area, which are used to find the probabilities of various transitions. This increases the accuracy of the forecast.

Further, at the determination step, the exit path can be determined, and the exit path contains a transition between the cell and the neighboring cell if the probability of exit through it is higher than another probability of exit. The determined exit path may be the path for which the probability of exit is highest. Thus, the motion recommendation for the driver of the vehicle can be pre-formed at a more likely exit from an open area, which means that it can be used faster.

In an additional embodiment of the present invention, at the step of dividing at least one subsequent transition from a neighboring cell to a next neighboring cell between the location and the edge of the region, a subsequent initial transition probability, or initial probability of a subsequent transition, can be assigned. At the correction step, the subsequent initial transition probability can be corrected if there is at least one type of the following information from the digital map related to the neighboring cell or the cell following it, with obtaining the adjusted subsequent transition probability. In addition, in the determination step, the probability of an exit can be determined using the adjusted subsequent probability for the subsequent transition. The next cell may be the fourth cell. The subsequent initial transition probability may be the third initial transition probability. The following information may relate to the second or fourth cell. The adjusted subsequent transition probability may be the third adjusted transition probability. The probability of exit can be determined using the first and third adjusted probability of transition. A similar embodiment of the present invention provides the advantage that it uses a sequence, or series, of transitions to determine the probability of leaving the area. This makes it possible to take into account the objects or phenomena of space contained in the cells that are further from the originally considered cell. Thereby, the reliability of the forecast with respect to where the abandonment of the open area will occur is further enhanced.

Brief Description of the Drawings

Below the invention is considered in more detail by examples of its implementation, illustrated by the accompanying drawings, which show:

figure 1 is a block diagram of a first embodiment of the present invention;

figure 2 is a block diagram of a second embodiment of the present invention;

figure 3 is a structural diagram of an embodiment of the present invention;

Fig. 4 is a diagram for explaining the steps to be taken in determining the probability of an exit in one embodiment of the present invention; and

5 is a fragment of a diagram explaining the steps performed in determining the probability of an exit in one embodiment of the present invention.

The implementation of the invention

The same and equivalent elements may be indicated in the drawings by the same or similar reference numbers, with no repeated description of such elements. In addition, in the drawings, in the description thereto, as well as in the claims, numerous features are contained in a particular combination. It is clear to the specialist that these features can also be considered separately or they can be combined in other combinations that are not explicitly described in this description. In addition, the essence of the invention in the description below may be disclosed using various sizes and dimensions, however, the possibilities of carrying out the invention are not limited to these sizes and dimensions. In addition, the steps proposed in the invention method can be repeated, as well as performed in a sequence different from that considered. If in one or another embodiment of the invention between the first sign / step and the second sign / step there is a compound union "and / or", this should be interpreted in such a way that in one embodiment of the invention both the first sign / step and the second sign are present / step and in another embodiment of the invention there is either only the first sign / step, or only the second sign / step.

Figure 1, as an explanation of the first embodiment of the present invention, shows a block diagram of the simulation of movements in the off-road area to predict future movements through the Markov process. The method for determining at least one probability of leaving the area identified in the digital map as an open area includes a partition step 110, a correction step 120, and also a determination step 130. In step 110 of the partition, the region is divided into a plurality of cells. At least one transition from one cell to an adjacent cell is assigned an initial probability of transition. Moreover, the indicated transition is between the location located inside this region and the edge of the region. At a correction step 120, the initial transition probability is adjusted to obtain the adjusted transition probability. Correction occurs when at least one type of information is obtained from a digital card. Moreover, such information refers to a cell and / or a cell adjacent to it. In a determination step 130, the probability of an exit is determined using the adjusted transition probability.

Proposed in the invention method, it is possible to determine the probabilities with which the driver, who got into his off-road area on his vehicle, will be able to get out of this area at one point or another on a digital roadmap. This allows you to determine among these points the point with the highest probability. To do this, a grid is applied to the off-road area in which the driver is on his vehicle. Grid fields serve as states for the Markov process. Using the topography of the area, the current location and orientation of the vehicle, and the vehicle model for physically correct motion simulation, as well as the characteristics of roads that limit the non-road area, the probabilities of transition from one state to another are calculated. Additionally, these probabilities can also be adjusted for the entry area. So, it is possible to calculate the probability of reaching an external state, which means that the vehicle will reach the road from its current location. Thus, the navigation system can quickly navigate when the vehicle returns to the road network and, if necessary, can begin to guide the actions of the driver, i.e. navigational support for vehicle movement. The probabilities thus determined are updated, for example, when the location of the vehicle changes, in particular with each such change. In addition, the method of the invention always offers some solution, even if the available data does not provide all the necessary information. Proposed in the invention method makes it possible to faster orientation when returning to the road network from an off-road area.

Figure 2 shows a block diagram of the implementation of the proposed method in another embodiment of the present invention. The method for determining at least one probability of leaving the area identified in the digital map as an open area is initialized in step 200. The location of the vehicle is indicated as being in an open area or off-road (Off-Road).

Step 110 partitioning the area is divided into several substeps. At a boundary identification substep 202, the boundaries of the off-road region are identified. In substep 204, overlay the area with the grid. In substep 206, determining the initial transition probabilities determines the initial transition probabilities. The correction step 120 is also divided into several substeps. In substep 208 of the refinement of the probabilities, the initial probabilities are adjusted taking into account the characteristics of the roads bordering the area. At substep 210, refinements taking into account terrain features initial probabilities are specified taking into account terrain characteristics. In substitute step 212, in the vicinity of the vehicle location, the aforementioned probabilities are specified taking into account physically possible movements of the vehicle. Additionally, step 214 of increasing the probability may be performed taking into account the entry point into the off-road area. At step 216, the probabilities are updated when the location changes.

The determination step 130 also includes several substeps. In substep 218, the determinations determine the result of calculating the probabilities of reaching the cells along the edges of the region. In substep 220, the most likely exit point is determined. As an alternative to this substep at substep 222, a normalized probability picture can be determined.

During the trip, the driver may leave the road network in his vehicle. However, most navigation devices are configured to support the driver with respect to navigational support for his movement to the destination along the roads provided for this, laid down in a digital map. Due to this functionality, the navigation system tries to correlate the driver’s location with a specific location on the road using the so-called map-binding method. When leaving a road, the system also attempts to map the location of the vehicle to the road. And only in case of going beyond a certain limit does the system cease to assume that the driver with his vehicle is still on the road. Thus, the location of the vehicle is no longer matched with roads. Instead, the system indicates the vehicle’s installed location in the off-road area. Usually, after a certain time, the driver again gets to the road network and continues to move along it. In the case of this repeated transient, the navigation system needs some time to adapt in order to find its location on the road and resume performing its main task, namely, managing the driver’s actions on the route. The method proposed in the invention improves this process by allowing the system to pre-calculate the possible and probable exit points from the off-road area and thus more quickly navigate if the driver leaves the off-road area. The method involves the use of available information to calculate at what point the driver on his vehicle is most likely to leave the off-road area. In this case, after initialization, the transition probabilities are sequentially refined with each receipt of information.

In the following description, an embodiment of the present invention corresponding to the block diagram of FIG. 2 is described in more detail. The method is initialized (step 200) if the current location is classified as off-road. For this, at the first step (step 202), the edges formed by the roads are defined, which limit the off-road area in which the driver is on his vehicle. If this area is too large or if due to its topographic features, for example, when it comes to the sea, it is not limited only to roads, then the size of the area is limited to a certain size, for example, one square kilometer.

If the driver, moving on his vehicle in an off-road area, reaches a similar artificially drawn (to limit the size) border, the method is re-initialized with a new area. This limits complexity by avoiding an excessive increase in computation time. A grid is also imposed on this area (step 204), including bounding roads. The description refers to a rectangular grid (raster), but at the same time other grids can be used. Each grid field serves as a state of a discrete Markov process. To implement such a Markov chain, it is required (step 206) the probability of transition, from one moment in time to another, to another state or staying in it. The sum of the probabilities as prescribed must be one. Then these probabilities are used to calculate the entire chain. Cells at the edge of the district receive the probability that the Markov chain will remain in this state equal to one. The probabilities that a transition to another cell will take place are set to zero. Thus, these cells represent the so-called stationary state in which the circuit remains.

In the remaining cells, the probability that the state in the cell is preserved is set to zero. The probability of a transition to a neighboring cell is initialized as uniformly distributed. Thus, in the case of a rectangular grid in which diagonal transitions are allowed, the probability of transition to the neighboring cell is 1/8. Thus, the initial transition probabilities are specified for all cells.

These probabilities are further refined using the available information. In the first step of refinement (step 208), for example, the characteristics of roads stored in a digital map are checked for how likely it is to get into them from an impassable area. So, the likelihood of getting on higher class roads, for example on expressways, is clearly lower than on smaller classes, for example, gravel roads. Such roads are usually digitized only insufficiently or directly lead to the off-road area. Therefore, the probability of leaving the non-road area on such a road is much higher.

At this step, the transition probabilities are adjusted for cells adjacent to external cells. At the same time, the probabilities of transition to all external cells are increased or decreased and, accordingly, the probabilities of transition to others located further inland of the cell region are also uniformly increased or decreased. Also, there is no change in the probability for external cells that are not part of the area restricting the road and part of the region’s artificially drawn border. With further implementation of the method, they are considered, for example, as being on an imaginary road. And only during the assessment they are considered differentially.

At the second possible refinement step (step 210), to correct the transition probabilities for the cells located within the region, information, if available, on the topography of the off-road region is used. If the area includes absolutely insurmountable places, such as lakes, then the probabilities of transition to the corresponding cells are set equal to zero. Altitudes are also used, which are taken into account in such a way that the transition probabilities for such (with a significant elevation difference) cells noticeably decrease, and for cells with a small elevation difference, they increase accordingly. The essence of this change is to assign higher probabilities to easier segments of the journey.

A third possible refinement (step 212) concerns cells located in the vicinity of the current location of the vehicle. In this case, the probabilities are adjusted in accordance with the physical conditions that determine what the movement of the vehicle may be. So, transitions from one cell to another, possible due to only unnatural turns, are assigned only very small probabilities, while the probabilities of transitions to cells, achieved by the rectilinear movement of the vehicle and its easy turns, are estimated higher. This step increases the complexity of the method, since the transition probabilities cease to be rigid, but change at each step due to changing location and direction of movement, however, such a local correction has a great influence on the probability of reaching the final state. It is when approaching the edge of the region that the exit point crystallizes in this way faster.

As a possible option, in the fourth step (step 214), you can increase the probability in the vicinity of the entry point into the off-road area. Thus, the assumption is made that the driver intentionally only briefly stays in the off-road area and intends to leave him in the same place. This is taken into account by increasing the probability of passing the corresponding transition. While the driver is in the vicinity of the corresponding cell, his vehicle is considered oriented to the entry point. The stronger the driver moves away from the entry point, the less the influence of this increase becomes, which corresponds to the assumption that the driver has outlined a different exit point. If there are all transition probabilities, then for all external cells, using the Markov chain, the probabilities of reaching these cells from the current location can be calculated (step 218). This calculation is carried out, for example, using conventional mathematical methods. Also in this way it is possible to determine (step 220) the cell with the highest probability of its achievement. This cell is taken at a given time as the most likely exit point from the off-road area. Here, when it comes to a cell located on an imaginary road, a differentiated approach can be applied. So, either it can be assumed that the driver does not intend to leave the non-road area so far. Or, you can search for an external cell located on a real road and having the highest probability of reaching it. As an alternative, it is also possible to normalize the probabilities of reaching all external cells, thus obtaining a general probabilistic picture of the exit points (step 222). In this case, you can also take into account all external cells or only cells that are on a real road.

It should be noted here that the calculation of the probabilities of reaching an external cell is always possible due to initialization and without additional information. In this case, the method corresponds to the shortest path search method by which the exit point that is located at the smallest distance from the current location is calculated.

If during the next measurement of the vehicle’s location it turns out that this location has changed, as a result of which the driver is in another cell, which in this case represents the next time step, then the execution of the method, for example, in the third refinement step (step 212) can be resumed ( step 216). At the same time, the transition probabilities in the vicinity of the cell in which the vehicle is located are re-adjusted. Cells that are no longer directly surrounded by the aforementioned cell receive initial transition probabilities without regard to the vehicle model. For those cells that are in the vicinity of the vehicle, the correction is carried out in accordance with the model of the vehicle (for example, in relation to the maximum possible agility, and therefore the reachability of these cells).

The previously calculated probabilities can be stored in memory (storage device) to be reused if the cell falls out of the vicinity of the vehicle again. After that, the implementation of the method is resumed, as described above, and new probabilities are calculated. To simplify these new calculations, certain partial circuits related to external cells can be stored in memory. If they are not affected by the correction in the vicinity of the current location of the vehicle, then they continue to remain valid. The more partial circuits are stored in memory, the faster the implementation of the method, but also the more memory space is used.

Figure 3 shows a block diagram of one embodiment of the present invention. The vehicle has a navigation system 300 and is located at a specific location. A device 310 is connected to the navigation system 300 to determine at least one probability of exiting the area identified in the digital map as an open area. The device 310 comprises a node 312 for dividing a region into a plurality of cells. At node 312, at least one transition from one cell to an adjacent cell is assigned an initial probability of transition. The transition is between the location of the vehicle and the edge of the area. The device 310 also includes a node 314 for correcting the initial probability of transition. At node 314, the initial transition probability is corrected if at least one kind of information is obtained from the digital map. Moreover, this information refers to the aforementioned cell or a neighboring cell. Node 314 provides an adjusted transition probability. At node 316, to determine the probability of an exit, the probability of an exit is determined at least using the adjusted transition probability. Using the found exit probability, the navigation system 300 can begin route guidance even before the vehicle reaches the edge of the area.

Figure 4 shows the sequence of actions and the results of implementing the method used to calculate the most likely exit point from the off-road area in one embodiment of the present invention. The vehicle is located at a location 400 within area 402. Area 402 is identified in an digital map as an open area. Region 402 is surrounded by borders 404. Such borders 404 are, for example, communication lines, water barriers, development zones or impassable terrain. Area 402 is covered by a grid 406 to form separate cells. In this embodiment, a quadrangular, in particular square, mesh is used as the mesh 406. Each transition from one cell of the grid 406 to a neighboring cell is assigned, or assigned, the initial probability of the transition.

Then follows the correction of the initial transition probabilities. To do this, in the cells 408 of the grid 406, the initial transition probabilities are adjusted. At the edge of region 402, correction is performed in response to the presence of cartographic information, for example, if there is information about the class of the communication path contained in the cell or about the difficulty of the reservoir contained in the cell, area, or built-up area. At a distance from the edge of the district, the correction is performed according to the insurmountability of the reservoir contained in the cell, site of the area or built-up area. If cells 410 are unreachable, then these cells and transitions in them are assigned low adjusted transition probabilities, which reflect the unlikely event that these cells come from neighboring cells surrounding them.

Cells 412 in the vicinity of the vehicle at location 400 receive transition probabilities, adjusted for the orientation and speed of the vehicle at location 400. The distance from the input cell 414, in which the vehicle entered an open area, affects the probabilities of transitions between cells in the neighborhood so that re-entry into the input cell 414 is more likely — up to a certain distance from the location to the input cell 414 in the open — than a miss. From location 400, for example, an exit path 416 is determined. The exit path 416 has an exit probability determined based on the initial transition probabilities and the adjusted transition probabilities, representing the probability of reaching a particular cell at the edge of the area. Thus, the exit path passes through several cells of an open area, divided by a grid. Thus, comparing, for example, the total transition probabilities obtained for individual cells through which several different exit paths leading to different exit cells at the edge of the region pass, it is possible to determine the exit path that has the highest total exit probability. This exit path describes the most likely path at a given time along which the vehicle will move from location 400. This exit path 416 leads to the cell located at the edge of the area, which is the most likely exit cell 418. Starting from exit cell 418, the navigation system in the vehicle will be able, even before the vehicle reaches exit cell 418, to prepare the process of navigating along the roads or proceed with this navigation, so that when the vehicle leaves an open area, no exit will occur in this exit cell delays.

5 shows a flowchart for determining the most likely exit cell in one embodiment of the present invention. There are several cells between the location and the edge of the area. At least one transition between cells, in this case, a transition between cell 500 and an adjacent cell 502, is assigned the initial probability of transition. If there is information received from the digital map, the initial probability of transition is adjusted to obtain the adjusted probability of transition from cell 500 to neighboring cell 502. The information contained in the digital map refers either to cell 500 or to neighboring cell 502. For example, this information may be an indication of that there is a lake or a pond in cell 502 that cannot be passed through. Using the adjusted transition probability, the probability of the vehicle in cell 500 entering the cell at the edge of the area is determined.

Another transition, in this case, a transition between cell 500 and another cell 504, is assigned a different initial transition probability. If the digital map has other information related either to cell 500 or to another cell 504, the other initial transition probability is adjusted accordingly. As a result, an adjusted other transition probability, i.e. the adjusted probability of transition between cell 500 and another cell 504. Using the adjusted different probability of transition between cell 500 and another cell 504, the probability of leaving the above location through another cell 504 located at the edge of the area is determined. Another cell 504 at the edge of the area may be a cell located at the edge of the area.

If the exit probability is higher than another exit probability, then an exit path is determined containing the transition probability (i.e., the transition between cells 500 and 502). If the other exit probability is higher than the exit probability, then another exit path is determined containing a different transition probability, i.e. the transition from cell 500 to another cell 504. Then, based on the sequence of transitions between adjacent cells, the exit path is determined, and to distinguish between individual transitions, the transition between cell 500 and the next cell 506 is called the next transition. Then, a subsequent initial transition probability is assigned to the subsequent transition between adjacent cell 502 and the next cell 506. If there is the following information obtained from the digital card related to the neighboring cell 502 and / or the next cell 506 following it, the subsequent initial transition probability is adjusted accordingly. As a result, an adjusted subsequent probability of transition between an adjacent cell 502 and a subsequent cell 506 is obtained.

Claims (9)

1. The method of determining at least one probability of leaving the area identified in the digital map as an open area (402), based on the location located within the area (400), comprising the following steps:
- dividing (110) a region into a plurality of cells (406) with assigning at least one transition between one location and the edge of the region from one cell (500) to the neighboring cell (502) of the initial transition probability;
- correction (120) of the initial transition probability, performed in the presence of at least one kind of information from a digital map relating to a cell or an adjacent cell, with obtaining an adjusted transition probability; and
- determination (130) of the probability of exit, performed at least using the adjusted probability of transition. 2. The method according to claim 1, wherein at the step of partitioning (110), the partitioning is performed in such a way that several transitions form a network, the network covering a region (402), and / or cells limited to a predetermined maximum area of the region. 3. The method according to claim 1, in which, at the correction step (120), the initial transition probability is adjusted using the class of the message path (404), if the cell (500) includes at least this message path, and the digital map contains information about the communication path and its class, and / or in which the initial probability of transition is adjusted using at least one terrain characteristic in cell (500) or in an adjacent cell (502) if the digital map contains information about the terrain characteristic. 4. The method according to claim 1, in which at the correction step (120), the initial probability of transition is adjusted using the direction of movement and / or speed of the vehicle in which the method is carried out. 5. The method according to claim 1, in which, at the correction step (120), the initial probability of the transition is adjusted using the transition distance to the point (414) of the vehicle’s entrance to the open area (402). 6. The method according to claim 1, in which:
- at the step of splitting (110), at least one other transition between the cell (500) and the edge of the region from the cell (500) to another neighboring cell (504) is assigned a different initial transition probability;
- at the correction step (120), another initial transition probability is corrected if there is at least one other type of information from the digital map related to cell (500) or another neighboring cell (504), to obtain an adjusted different transition probability; and
- in the determination step (130), another exit probability is determined, the other exit probability being determined at least using the adjusted different transition probability. 7. The method according to claim 6, in which, at the step of determining (130), a path (416) for exiting the region (402) is determined, the exit path comprising a transition between cell (500) and an adjacent cell (502), if the probability of exit through it higher than another exit probability. 8. The method according to one of the preceding paragraphs, in which
- at the step of partitioning (110), at least one region located between the location (400) and the edge (402) of the subsequent transition from the neighboring cell (502) to the next neighboring cell (506) is assigned the subsequent initial transition probability;
- at the correction step (120), the subsequent initial transition probability is corrected if there is at least one type of the following information from the digital map related to the neighboring cell (502) or the next cell (506), with obtaining the adjusted subsequent transition probability; and
- in the determination step (130), the exit probability is determined using the adjusted subsequent probability for the subsequent transition. 9. A device (310) for determining at least one probability of leaving the area identified in the digital map as an open area (402), based on a location (400) located within the area, containing:
- a node (312) for dividing a region into a plurality of cells (406) with assigning at least one transition located between the location and the edge of the region from one cell (500) to the neighboring cell (502) of the initial transition probability;
- a node (314) for correcting the initial probability of the specified transition, performed when there is at least one type of information from the digital map related to the specified cell (500) or its neighboring cell (502), with obtaining the adjusted transition probability; and
- a node (316) for determining an exit probability, which is performed at least using the adjusted transition probability.

Images